Biomedical Engineering Reference
In-Depth Information
Figure 22.3 Data sources can be segregated according to particular data slices at
deposition. Each chemical record displays the types of associated data sources under
a set of tabs. Where possible the external IDs associated with each data source are
linked out to the website associated with the data source. The tabs displayed are associ-
ated with the chemical record for alprazolam [http://www.chemspider.com/2034].
molecular weight, nominal, average, and monoisotopic masses, isomeric simpli-
fi ed molecular input line entry specifi cation (SMILES) string [60], InChIString,
and InChiKey [61]. The chemicals are also passed through two separate prop-
erty prediction suites, the ACD/Labs [62] and EPISuite [63] programs, to
generate physicochemical properties such as log P , boiling point, and many
others. While not exhaustive in terms of the general applicability of the algo-
rithms to the entire structure space, for small organic molecules these algo-
rithms generally provide excellent predicted values that can be used as fair
estimates and good fi lters during searching that may be useful for understand-
ing “lead or druglike” properties. As a result of the added value provided to
every chemical record at deposition, every record can be a rich contributor to
the overall data set that can then be used by scientists as they see fi t.
Many of the online databases are focused in particular areas. For example,
the Human Metabolome Database [15] is concerned specifi cally with molecule
metabolites found in the human body while a number of the depositions are
those of chemical vendors listed with an intention to provide access to their
compound collection to the community and generate business opportunities.
ChemSpider has segregated deposited data into slices using labels including
chemical vendors, biological data, metabolism data, natural products, and so
on, as shown in Figure 22.3. At the time of deposition of the data associated
with a data source the particular segregation fl ags are defi ned by the data
provider. There are no limits to the number of data slices with which a particu-
lar depositor can be associated. In the interface viewed by the user the data
sources are displayed under tabs. The relevant data sources are listed together
with the associated external identifi er (generally a number) and, where pos-
sible, with a direct Uniform Resource Locator (URL) linking the external ID
into the data source website. In this way a user can quickly navigate to infor-
mation hosted on external sites, a feature which is particularly important when
Search WWH ::




Custom Search